{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T22:11:45Z","timestamp":1793052705649,"version":"4.5.0"},"reference-count":27,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T00:00:00Z","timestamp":1774569600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Digit. Health"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Diabetes is a chronic disease characterized by elevated blood glucose levels. Without early detection and proper management, it can lead to serious complications and increase healthcare costs. Its global prevalence is rising, with many cases remaining undiagnosed. In this study, we developed an explainable machine learning model using a two-stage approach for predicting diabetes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Five machine learning (ML) models, including Multi-Layer Perceptron, Support Vector Machine, K-Nearest Neighbor, Extreme Gradient Boosting (XGBoost), and Na\u00efve Bayes, were trained and evaluated using a two-stage approach. In Stage one, a public dataset containing 520 samples was used, and Shapley Additive exPlanations (SHAP) and MLP weights were applied for feature selection. In Stage two, the same models were trained and evaluated using a dataset of 270,943 samples collected from Rwanda. SHAP was further employed to explain the model output.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In Stage one, the Multi-Layer Perceptron model achieved the best performance on a public dataset, with an accuracy of 95.19%. Feature selection techniques identified the top 10 influential predictors associated with diabetes risk, including those recommended by diabetes care providers in Rwanda. In Stage two, the XGB model outperformed other models, achieving an accuracy of 97.14%.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This study presents a two-stage, explainable machine learning framework for systematic screening for type 2 diabetes. The first stage evaluates risk based on reported symptoms, while the second stage incorporates demographic, anthropometric, and vital sign data for refined risk assessment. Integration of these models into the mUzima mobile application can enhance community health workers' capacity to identify and refer high-risk individuals. By enabling early and accurate detection, the proposed approach has the potential to reduce undiagnosed diabetes and support improved disease management.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1743619","type":"journal-article","created":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T05:19:35Z","timestamp":1774588775000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Early Type 2 diabetes risk prediction using explainable machine learning in a two-stage approach"],"prefix":"10.3389","volume":"8","author":[{"given":"Silas","family":"Majyambere","sequence":"first","affiliation":[{"name":"Department of Computer and Systems Sciences, Stockholm University","place":["Stockholm, Sweden"]},{"name":"Centre of Excellence in Biomedical Engineering and E-Health (CEBE), University of Rwanda","place":["Kigali, Rwanda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tony","family":"Lindgren","sequence":"additional","affiliation":[{"name":"Department of Computer and Systems Sciences, Stockholm University","place":["Stockholm, Sweden"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Celestin","family":"Twizere","sequence":"additional","affiliation":[{"name":"Centre of Excellence in Biomedical Engineering and E-Health (CEBE), University of Rwanda","place":["Kigali, Rwanda"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Isaac","family":"Ntakirutimana","sequence":"additional","affiliation":[{"name":"Centre of Excellence in Biomedical Engineering and E-Health (CEBE), University of Rwanda","place":["Kigali, Rwanda"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,3,27]]},"reference":[{"key":"B1","volume-title":"IDF Diabetes Atlas","year":"2025"},{"key":"B2","doi-asserted-by":"publisher","first-page":"5641","DOI":"10.3390\/ijerph20095641","article-title":"Community health worker-led cardiovascular disease risk screening and referral for care and further management in rural and urban communities in Rwanda","volume":"20","author":"Niyibizi","year":"2023","journal-title":"Int J Environ Res Public Health"},{"key":"B3","doi-asserted-by":"crossref","first-page":"459","DOI":"10.3233\/SHTI240447","article-title":"Muzima Mobile application for screening hypertension and diabetes: a user experience of the app among community health workers in Rwanda","volume-title":"Digital Health and Informatics Innovations for Sustainable Health Care Systems","author":"Mugisha","year":"2024"},{"key":"B4","doi-asserted-by":"publisher","first-page":"115","DOI":"10.11604\/pamj.2022.41.115.30682","article-title":"Prevalence and characteristics associated with diabetes mellitus and impaired fasting glucose among people aged 15 to 64 years in rural and urban Rwanda: secondary data analysis of world health organization surveillance data","volume":"41","author":"Bavuma","year":"2022","journal-title":"Pan Afr Med J"},{"key":"B5","doi-asserted-by":"crossref","DOI":"10.1109\/ICCCI54379.2022.9740844","article-title":"Type-II diabetes prediction using machine learning algorithms","author":"Charitha","year":"2022"},{"key":"B6","doi-asserted-by":"crossref","DOI":"10.1109\/PDGC56933.2022.10053092","article-title":"Hybrid prediction model for type-2 diabetes Mellitus using machine learning approach","author":"Bhat","year":"2022"},{"key":"B7","doi-asserted-by":"crossref","DOI":"10.1109\/ETNCC63262.2024.10767487","article-title":"Beyond glucose levels: a machine learning perspective on type 2 diabetes prediction","author":"Aslam","year":"2024"},{"key":"B8","doi-asserted-by":"publisher","first-page":"37370","DOI":"10.1109\/ACCESS.2024.3422319","article-title":"A comparative analysis of LIME and SHAP interpreters with explainable ML-based diabetes predictions","volume":"13","author":"Ahmed","year":"2024","journal-title":"IEEE Access"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1049\/htl2.12039","article-title":"Diabetes prediction using machine learning and explainable AI techniques","volume":"10","author":"Tasin","year":"2023","journal-title":"Healthc Technol Lett"},{"key":"B10","doi-asserted-by":"publisher","first-page":"103069","DOI":"10.1016\/j.eclinm.2025.103069","article-title":"Prediction model for type 2 diabetes mellitus and its association with mortality using machine learning in three independent cohorts from South Korea, Japan, and the UK: a model development and validation study","volume":"80","author":"Lee","year":"2025","journal-title":"eClinicalMedicine"},{"key":"B11","doi-asserted-by":"publisher","first-page":"180","DOI":"10.1186\/s12902-020-00660-y","article-title":"Socio-demographic and clinical characteristics of diabetes mellitus in rural Rwanda: time to contextualize the interventions? A cross-sectional study","volume":"20","author":"Charlotte","year":"2020","journal-title":"BMC Endocr Disord"},{"key":"B12","doi-asserted-by":"publisher","first-page":"34163","DOI":"10.1007\/s11042-023-15165-8","article-title":"Efficient prediction of early-stage diabetes using XGBoost classifier with random forest feature selection technique","volume":"82","author":"G\u00fcndo\u011fdu","year":"2023","journal-title":"Multimed Tools Appl"},{"key":"B13","first-page":"1","article-title":"Enhancing interpretability in diabetic prediction: a comparative study of SHAP, LIME, and permutation feature importance","volume-title":"AI-Driven Smart Healthcare for Society","author":"Uddalak","year":"2025"},{"key":"B14","doi-asserted-by":"crossref","DOI":"10.1109\/INCOFT60753.2023.10425305","article-title":"Early-Stage diabetes risk prediction using supervised machine learning algorithms","author":"Islam","year":"2023"},{"key":"B15","doi-asserted-by":"publisher","first-page":"19462","DOI":"10.1007\/s11227-024-06211-9","article-title":"A novel hybrid deep learning model for early stage diabetes risk prediction","volume":"80","author":"B\u00fclb\u00fcl1","year":"2024","journal-title":"Journal of Supercomputing"},{"key":"B16","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1108\/ACI-01-2021-0022","article-title":"Classification models for likelihood prediction of diabetes at early stage using feature selection","volume":"20","author":"Oladimeji","year":"2024","journal-title":"Applied Computing and Informatics"},{"key":"B17","article-title":"Early Classification of Diabetes Dataset","year":""},{"key":"B18","doi-asserted-by":"publisher","first-page":"863","DOI":"10.1613\/jair.1.11192","article-title":"SMOTE For learning from imbalanced data: progress and challenges, marking the 15-year anniversary","volume":"61","author":"Fernandez","year":"2018","journal-title":"J Art Intell Res"},{"key":"B19","doi-asserted-by":"publisher","first-page":"3926","DOI":"10.4108\/eetpht.9.3926","article-title":"Disease prediction using a modified multi-layer perceptron algorithm in diabetes","volume":"9","author":"Dayal","year":"2023","journal-title":"EAI Endorsed Trans Perv Health Technol"},{"key":"B20","doi-asserted-by":"publisher","first-page":"1499530","DOI":"10.3389\/frai.2024.1499530","article-title":"Robust predictive framework for diabetes classification using optimized machine learning on imbalanced datasets","volume":"7","author":"Abousaber","year":"2025","journal-title":"Front Art Intell"},{"key":"B21","doi-asserted-by":"publisher","first-page":"171","DOI":"10.14569\/IJACSA.2024.0150818","article-title":"Diabetes prediction using machine learning with feature engineering and hyperparameter tuning","volume":"15","author":"Massari","year":"2024","journal-title":"J Adv Comput Sci Appl"},{"key":"B22","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1186\/1472-6947-10-16","article-title":"Application of support vector machine modeling for prediction of common diseases: the case of diabetes and pre-diabetes","volume":"10","author":"Yu","year":"2010","journal-title":"BMC Med Inform Decis Mak"},{"key":"B23","first-page":"102","article-title":"Real time diabetes prediction using na\u00efve Bayes classifier on big data of healthcare","volume":"7","author":"Shah","year":"2020","journal-title":"Int Res J Eng Technol (IRJET)"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2012.01536","article-title":"Improving KernelSHAP: practical shapley value estimation via linear regression","author":"Covert","year":"2021","journal-title":"(AISTATS)"},{"key":"B25","doi-asserted-by":"publisher","first-page":"34","DOI":"10.69882\/adba.cem.2024075","article-title":"Machine learning interpretability in diabetes risk assessment: a SHAP analysis","volume":"1","author":"Kutlu","year":"2024","journal-title":"Comput Eletron Med"},{"key":"B26","doi-asserted-by":"publisher","first-page":"3688","DOI":"10.1186\/s12889-025-24953-w","article-title":"Leveraging XGBoost and explainable AI for accurate prediction of type 2 diabetes","volume":"25","author":"Rafie","year":"2025","journal-title":"BMC Public Health"},{"key":"B27","doi-asserted-by":"publisher","first-page":"3738","DOI":"10.3390\/electronics14183738","article-title":"An explainable AI framework for online diabetes risk prediction with a personalized chatbot assistant","volume":"14","author":"Maimaitijiang","year":"2025","journal-title":"MDPI Electronics"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1743619\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T05:19:37Z","timestamp":1774588777000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1743619\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,27]]},"references-count":27,"alternative-id":["10.3389\/fdgth.2026.1743619"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1743619","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,27]]},"article-number":"1743619"}}